University of California, San Diego ISSN: 0888-73381

Center for Magnetic Recording Research RReeeporpoppoorrttt Number 19 WINTER • 2002

CONTENTS

1 Shannon Symposium & Statue Dedication Shannon Symposium and 3Director’s Message Statue Dedication at CMRR 4 Cal (IT)2 Workshop on Non-Volatile Memory

5CMRR Research Review

Research Highlights plied the resources to videotape the t 2 PM on October 16, symposium proceedings, as well as 6 Recent Gifts, Grants, 2001, a statue of Claude interviews with many of the speakers, Awards, and Internships Elwood Shannon, the for use in a pair of UCSD-TV docu- 7Honors and Awards A Father of Information mentaries highlighting the enormous Theory, was dedicated in the lobby of impact of Shannon’s genius. Ph.D.’s Awarded the Center for Magnetic Recording The acquisition of the sculpture Research (CMRR) at the University and the organization of the Shannon 8 New Visitors of California, San Diego. In conjunc- Symposium were spearheaded by 9 UCSD Open House tion with the dedication, a sympo- Professor Jack K. Wolf, a member sium honoring Shannon’s life and of the Electrical and Computer 10 Guzik Tester work was held on October 15th Engineering department in the and the morning of October 16th. Jacobs School of Engineering, and 11 Obituary for The symposium and dedication holder of one of the four CMRR Peter McEwen ceremony were attended by approxi- endowed chairs. 12 Current CMRR Sponsors mately 100 persons, many of them The Shannon Symposium program UCSD students. included invited technical presenta- Generous sup- tions and personal reflections by port for both fourteen information theorists from events, including industry and academia, including a live webcast of CMRR Director Paul Siegel and the symposium, UCSD Professor . was provided by Many of the speakers themselves the California have made enormous contributions Institute of Tele- to the astounding advances in communications telecommunications that have trans- and Information formed our world in the past half- Technology century. Among them were six recipi- (Cal(IT)2) at UC, ents of the prestigious Claude E. San Diego. The Shannon Award, the highest techni- Jacobs School of cal honor bestowed upon an individ- OCTOBER 16, 2001, SHANNON STATUE UNVEILING. LEFT TO RIGHT: DEAN ROBERT CONN, SCULPTOR EUGENE DAUB, SHANNON STATUE, DR. IRWIN Engineering and ual by the IEEE JACOBS, VICE CHANCELLOR MARSHA CHANDLER, HOST UCSD-TV sup- Society. The Shannon Award, which continued on page 2

CMRR REPORT1 WINTER•2002 Shannon Symposium and Statue Dedication at CMRR continued from page 1

Prof. Wolf himself received in 2001, The dedication ceremony that fol- The bronze plaque on the pedestal is given annually in recognition lowed the Symposium, hosted by Jack of this statue reads: of consistent and profound contribu- Wolf, included remarks from UCSD tions to the field of information Senior Vice Chancellor Marsha CLAUDE ELWOOD SHANNON theory. Chandler; UCSD Professor and 1916-2001, The Symposium speakers and their Director of the San Diego Division of Father of Information Theory presentation titles were as follows: the Cal(IT)2 Ramesh Rao; and His formulation of the mathematical Jacobs School of theory of communication provided Engineering Dean the foundation for the development , “Living Systems are Shannon Robert Conn. The of data storage and transmission Optimum Without Coding” next speaker, systems that launched the Paul Siegel, “The Continuing Miracle of Qualcomm Chairman information age. Information Storage Technology” and CEO - and former Dedicated October 16, 2001, UCSD faculty mem- Eugene Daub, Sculptor and Alon Orlitsky , “Universal Data ber - Dr. Irwin Jacobs, Compression” commented on the As depicted in the sculpture, David Neuhoff, “Time Stamp Coding-A Problem enormous impact of Shannon holds in his left hand a Shannon Did Not Answer” Shannon’s work upon sheet of paper. On this sheet is Thomas Cover, “The Value of State Information communications and inscribed a formula taken from in Communications and ” shared a fascinating Shannon’s landmark 1948 paper, historical tidbit, “A Mathematical Theory of G. David Forney Jr., “Approaching Channel namely that in 1967 Communication,” that is universally Capacity” was acknowledged to be the genesis of Edward van der Meulen, “The Duality Between named a Fellow of information theory. The formula, Successive Refinement of Information by Muir College at selected by Jack Wolf for its rele- Source Coding with a Fidelity Constraint and UCSD. Sculptor vance to both digital data recording Efficient Multilevel Channel Coding Eugene Daub con- and transmission, gives the capacity Under Cost Constraints” cluded with remarks of a discrete noisy channel: about the creative C = Max (H(x) - H (x)). Robert Lucky, “Impact of Shannon on Modern y process from which Be sure to enjoy the sculpture that Telecommunications” emerged his beautiful now graces the CMRR lobby on Ian Blake, “Randomness and Determination in and moving work of your next visit, and look for the ” art. Finally, to the symposium proceedings and further , “Quantized Iterative Decoding delight of all in atten- details about the UCSD-TV docu- with Closed-Form Density Evolution dance, the bust was mentaries on the CMRR and 2 Recursions for LDPC Codes on the AWGN unveiled by Rachel Cal(IT) websites. Channel” Wolf, Jack Wolf’s granddaughter. Solomon Golomb, “The Claude Shannon I Knew” , “Shannon’s Work on Block Code Performance and It’s Impact” CLAUDE ELWOOD SHANNON Shu Lin, “Construction of Low Density Parity Check Codes: Combinatorial Approaches” 1916-2001 Father of Information Theory Robert McEliece “The Generalized Distributive Law (with Loops) and Free Energy Minimization”

CMRR REPORT2 WINTER•2002 FROM THE DIRECTOR

Our Fiscal Year 2000-2001 research these activities, as well as profiles of budget was close to $2.5M, including some of our visiting students and contributions from our industrial researchers. You will also learn of partners, federal and state funding professional and technical recogni- agencies, and the University of tion received by our faculty, research California. We have supported collab- scientists, and students, including orations with affiliated UCSD faculty Jack Wolf’s selection by the IEEE in Computer Science and Information Theory Society as the Engineering (SMART - drive failure recipient of the prestigious 2001 prediction), Electrical and Computer Claude E. Shannon Award, and Engineering (coding for optical fiber UCSD senior Heidi Buck’s first prize channels for storage networks), in the ECE Department’s undergrad- Mechanical and Aerospace uate research conference for her Engineering (servo control), and CMRR research project. And there is Physics (magnetic semiconductors more that I believe will be of interest and magnetic nanoparticles). The and value to you, as well. National Security Agency (NSA) is In closing, let me thank you for sponsoring a particularly timely your interest in CMRR. Information research project addressing two seem- storage technology is a vital pillar - he CMRR newsletter is ingly opposite objectives - secure era- along with communications and back! The revival of the sure of information on disks and computing - in the evolving net- newsletter is part of the tapes, and recovery of information worked infrastructure that will be the T Center’s overall effort to from “erased” storage devices. foundation of our global society. The more effectively communicate with CMRR faculty and scientists are research and educational programs at our partners in industry and in gov- active participants in the National CMRR have a proud tradition, com- ernment agencies, to more actively Storage Industry Consortium (NSIC) bining technical innovation in mag- engage with our many alumni/ae, and research programs, with several proj- netic recording with a commitment to more clearly articulate to the gen- ects now being funded by the to produce creative, motivated, and eral public the importance of infor- Extremely High Density Recording skilled graduates who will accelerate mation storage technology in today’s (EHDR) “Terabit per square inch” the evolution of storage technology networked infrastructure. To this project and the TAPE project. The from positions in industry, govern- end, the newsletter will highlight Center is also playing an active role ment labs, and academia. important research developments, in, and benefiting from, the evolution The future holds both opportunity offer background on new technical of the California Institute for and challenge for CMRR, as it does initiatives, showcase students and Telecommunications and Information for the data storage industry as a researchers, and provide news for Technology, or Cal(IT)2, which whole. We are delighted to be in a and about CMRR alumni/ae. recently celebrated its first anniver- strong position to contribute to what This semi-annual publication is sary. The Institute sponsored a I like to refer to as the “continuing intended to complement the other Workshop on Non-Volatile Memory miracle of information storage tech- resources that CMRR provides to our (organized by CMRR Affiliated nology.” partners for information about the Faculty Professor Frances Hellman I hope that our corporate and gov- Center’s progress: the Spring and Fall of Physics), the Shannon Symposium ernmental sponsors will continue to Research Reviews, the CMRR and Statue Dedication (organized by recognize the enormous value - to Information Center, and the CMRR CMRR Professor Jack Wolf), and a them, the storage industry, and socie- Website. (The newsletter will also be Summer Undergraduate Research ty as a whole - that derives from their posted on our Website at Fellow. The Institute is also fostering investment in CMRR. http://www.ucsd.edu/cmrr) and providing support for evolving Enjoy the newsletter! During the past 18 months, the CMRR collaborations with Center has not only maintained its researchers in Physics and in the San strength in core research areas, but Diego Supercomputer Center. also expanded into new areas vital to In this issue of the newsletter, you the future of magnetic recording. will find articles about several of — Paul H. Siegel, Director

CMRR REPORT 3 WINTER•02 Cal (IT)2 Workshop on Non-Volatile Memory

rances Hellman, Professor UC Irvine and UC San Diego are will involve self-assembly, nanomag- of Physics at UCSD and being designed to enhance research in netism, and “spintronics” (spin elec- CMRR Affiliated Faculty key areas for materials and devices. tronics). Added Schuller: “If you look F member, organized and “Opto-electronics, quantum comput- 20 years into the future, you’re talk- hosted the Cal-(IT)2 Workshop on ing and communications, non-volatile ing about quantum magnetics, work- Non-Volatile Memory, held at CMRR, memories, and materials research are ing at the molecular level, and quan- September 21-22, 2001. The objec- all core competencies for Cal-(IT)2,” tum computing.” tive of the workshop was to develop a said Smarr. “Non-volatile memory has After Schuller’s presentation, plan for collaborative research, in a major role to play in a future where Dieter Weller, director of media partnership with industry, in the area billions of wireless sensors and other research at Pittsburgh-based Seagate of advanced non-volatile Technology, expand- memory technologies. A ed on “The Future blue-chip roster of of Magnetic speakers from leading “The institute recognizes the importance of put- Recording.” companies in the infor- “Magnetic recording ting academics and business researchers together. mation storage industry will not be dis- delivered stimulating It quickly became clear that there are a number lodged in the fore- technical presentations seeable future,” he to a group of more than of areas where our research can complement what said. “It is a very 50 academic and indus- low-cost method of trial researchers. Among industry is doing, especially on technologies that storing vast the speakers were repre- amounts of data sentatives of Seagate won’t be commercial for a few more years.” with almost no com- Technology and IBM, — Frances Hellman petition. Magnetic two of the founding recording is hard to sponsors of CMRR and compete with from both vital partners in a capacity perspec- collaborative research ever since. devices will be gathering, storing, tive as well as from a technology The audience included 14 faculty transmitting and even processing vast maturity perspective. At the moment, members from UC Irvine and UC San amounts of information.” recording densities of about 100 Diego, including 3 from CMRR, rep- UCSD Professor Ivan Schuller Gigabits per square inch have been resenting the Materials and Devices updated attendees on basic research demonstrated. We think we can push layer and the Networked issues related to the future of magnet- the technology up to perhaps 150-200 Infrastructure layer within Cal-(IT)2. ics, urging scientists to “follow your Gigabits per square inch using con- “The institute recognizes the impor- nose! Don’t let applications or imme- ventional longitudinal [technology], tance of putting academics and busi- diate business dictate where you look then we may move to perpendicular ness researchers together,” said for the answers.” Schuller also recording pushing toward 500 Hellman. “It quickly became clear advised researchers to “push limits”-to Gigabits per square inch or higher. At that there are a number of areas smaller sizes, more complex materials, terabit per square inch areal densities where our research can complement using state of the art technologies, a bit cell is only 25 nano-meters on what industry is doing, especially on and exploring unusual geometries. the side. We are eventually approach- technologies that won’t be commer- Looking out ten years, the leader of ing the point where we have single-, cial for a few more years.” Cal-(IT)2’s Materials as opposed to multiple-particle-per-bit In welcoming workshop partici- and Devices group at UCSD (his recording. At Seagate we are very pants, Cal-(IT)2 director Larry Smarr counterpart at UC Irvine is Professor interested in novel media approaches. said that the institute’s buildings at G.P. Li) said the technology frontier continued on page 10

CMRR REPORT4 WINTER•2002 CMRR Research Review

MRR hosted members of the CMRR Industrial Sponsor companies and special invited guests at Systems Program, San Diego meeting adjourned at 4:30 p.m., the semi-annual Research SuperComputer Center. however guests were invited to stay CReview and Advisory Council meet- Dr. Gordon Hughes discussed for optional lab tours, private dis- ing on October 17-18. The two days Intelligent Disk Drives, and Prof. Ami cussions, and the IEEE San Diego of talks highlighted the work of the Berkowitz's group ended the first Magnetics Society Chapter meeting. CMRR researchers and graduate stu- day's session with presentations on Guest speaker, Nick Rizzo, from dents. The Wednesday agenda Magnetic Materials and Devices. Late Motorola Labs, gave a presentation began with talks on Signal afternoon and evening activities entitled "Magnetoresistive Random Processing for Recording from Prof. included tours of the CMRR laborato- Access Memory - 256kb MRAM Jack Wolf’s group, followed by pre- ries, private discussions with students Wafer Progress.” sentations on Coding and and researchers, and dinner. The The next CMRR semi-annual Modulation by members of Prof. Advisory Council meeting followed. Research Review is scheduled Paul Siegel's group. The afternoon Day two began with presentations on for May 22-23, 2002. activities began with a special ses- Recording Physics and CMRR Industrial Sponsor company sion featuring a talk on Storage Micromagnetics by members of Prof. employees may contact Cheryl Hacker Systems and Networks by Chaitanya Neal Bertram's group, followed by (858-534-6563 or [email protected]) Baru and Phil Andrews, Program Co- talks on Tribology and Mechanics to ensure that your name appears on Directors, Data and Knowledge from Prof. Frank Talke's group. The our invitation list

Research Highlights

S.M.A.R.T. (Self-Monitoring and SMART technology, to assist them in minimize the number of good drives Reporting Technology) is a enhancing their failure warning returned, while maintaining maximum UCSD/CMRR research program to accuracy and warning time, while accuracy. improve the failure warning technolo- reducing false alarms from good Improved SMART prediction meth- gy currently built into disk drives. drives. Experimental SMART failure ods have been developed at CMRR Long term goals include drive reliabili- warning data is analyzed by data that achieve the maximum accuracy ty improvement by SMART technolo- mining and pattern recognition possible, at these low false alarm gy assessment of individual drives statistical techniques. rates. They are simple to implement in during their manufacturing final test, Significant opportunities exist for disk drive internal firmware. and by building expert systems for SMART enhancements. Its maximum computerized failure analysis. The failure warning accuracy is estimated program is described in the SMART to be about 30% overall in the disk web site at cmrr.ucsd.edu/smart and drive industry today. Disk drive field is sponsored by the UCSD return rates are about 1% per year, Information Storage Industry Center, and more than one-third of these are a Sloan Foundation Center. false alarms. The drive checks out The research program works with okay. The SMART false alarm rate disk drive manufacturers on their needs to be kept at 0.1%-0.2% to

CMRR REPORT5 WINTER•2002 Recent Gifts, Grants, Awards, and Internships

National Storage Industry Consor- Dr. Spada’s research involves the IBM has recently awarded Prof. Neal tium (NSIC) grant to Paul Siegel “Contribution of Electrochemical Bertram both a 2001 IBM Faculty as part of the Extremely High Processes to Increased Head- Partnership Award, and an IBM Density Recording (EHDR) 1 Media Spacing in Tape Drives.” It Fellowship Award. Terabit-per-square-inch program. is expected that the results of the The Faculty Partnership Award The project title is "Error Control studies will identify potential elec- involves computer simulations of Coding for Terabit per Square trochemical incompatibilities Ultra High Speed Magnetization Inch Recording." The research among the tape/environment and Reversal in Magnetic Recording addresses performance and imple- candidate head materials that can Heads. The research will develop mentation issues for enhanced contribute to increased magnetic a code that focuses on the head error-correction coding schemes, spacing during operation. tip region where the recording particularly soft-decision Reed- field emanates. The numerical Prof. Ami Berkowitz has recently Solomon decoding algorithms. problem involves incorporating received funding from the NSIC grant to Neal Bertram, Paul intrinsic magnetization dynamics Lawrence Livermore National Siegel and Jack Wolf as part of with the effects of the material Laboratory, IAP Research, Inc., the EHDR 1 Terabit-per-square- conductivity. and a Multi University Research inch program. The project title is Initiative (MURI) funded by the The Fellowship Award supports Zhen "Software Development of a Data ONR via the Massachusetts Jin, a Bertram graduate student Stream Simulator for Institute of Technology. researcher for the 2001-2002 Perpendicular Recording at nine-month academic year. Mary Densities of 300 Gbit-per-square- The Livermore project’s objective is Doerner will act as the technical inch and Beyond." The purpose of to eliminate chard formation dur- liaison and main point of contact the study is to develop a data sim- ing the spark erosion of Terfenol. at IBM. The fellowship will sup- ulator that can rapidly generate Prof Berkowitz’s group will inves- port Zhen’s research on the role large streams of playback pulses tigate two approaches to elimi- of medium microstructural design while incorporating realistic physi- nate this problem. The first is to on system performance. cal models of the perpendicular produce very short (<1 millisec- recording process. ond) pulses for the spark erosion The National Security Agency in order to drastically reduce the has awarded a contract to Prof. Frank Talke and Dr. Fred strained area. The second Dr. Gordon Hughes for a continu- Spada have each recently approach is to use the chards ing disk erasure study. The project received continued funding from and/or small Terfenol particles will support the implementation NSIC under their Program in formed by milling and subject of the “Security Erase” technolo- Advanced Magnetic Tape Storage them to arcing in argon in a con- gy that CMRR has developed, Technology. Prof. Talke’s research fined Delrin cell with 60 Hz power whereby disk drives containing centers on the “Investigation of source. This process is referred to confidential material can be Tape Edge Wear,” which address- as “refrying” the Terfenol. erased in a manner assuring com- es the characteristics of tape edge plete destruction of defects and tape edge wear as a The IAP Research project calls for all prior recorded information function of tape velocity, tape lat- the selection of appropriate mag- (“shredding electronic data”). eral guide force, local contact net powder compositions suitable After a drive reports a successful pressure, mechanical properties of for high frequency applications Security Erase (“SE”) completion, the guide pad, and thickness of (<100 MHz) and preparation of there should be no technique that tape substrate and tape magnetic powders using a spark erosion would permit recovery of the coating. process. information. Prof. Berkowitz’s MURI project coordinates the magnetic and structural characterization of spark-eroded Ferromagnetic Shape Memory Alloy particles.

CMRR REPORT6 WINTER•2002 HONORS Ph.D.’s Awarded AND Two former graduate students in Prof. Neal AWARDS Bertram’s research group have recently been awarded their Ph.D. degrees.

June 2001 Hong Zhou’s Ph.D. research stud- Prof. Jack Keil Wolf, recipient of ies involved performing large-scale the 2001 Claude E. Shannon Award numerical simulations of the mag- of the IEEE Information Theory netic recording process. The focus Society, delivered the Shannon was the effect of granular interac- Lecture at the 2001 IEEE tions, both exchange and magneto- International Symposium on static, on signals, noise, and ther- Information Theory, which was held mal decay. He studied longitudinal in Washington, D.C., June 24-29, (current media), both planar ran- 2001. The title of his lecture was dom and oriented, and proposed perpendicular recording for densi- "Constrained Sequences, Crossword 2 Puzzles and Shannon." The Claude ties above 100Gbit/in . He found E. Shannon award is the highest optimal medium magnetic parame- honor bestowed by the IEEE ters for these three types of Information Theory Society in recording. After leaving CMRR, recognition of consistent and pro- Hong was employed by Seagate found contributions to the field of Research in Pittsburgh. information theory.

March 2001 Alfred Liu completed his Ph.D. Heidi Buck, a senior in Prof. Jack degree in December 2000 and Wolf’s ECE 155B Digital Recording continued as a postdoc with Prof. Projects class, Winter Quarter 2001, Bertram until February 2001. He shared first prize at the ECE did a TA ship with UCSD’s Undergraduate Research Physics Department until June Konference and Assembly (EURE- 2001. Alfred focused during his KA), which was held at CMRR, Ph.D. studies on micromagnetic March 16, 2001. Her project was effects in thin films, including the entitled "Investigation of Channel effects of finite temperature. He Noise Variance Estimation on Turbo applied these models to explain Decoder Performance." measurements of experimental groups at UCSD. The areas were January 2002 switching mechanisms in multilay- ered soft thin films and tempera- The IEEE Fellow Committee has ture effects in thin granular films. named Senior Member James U. He concluded his thesis with a Lemke to Fellow Grade effective theoretical study to explain details January 1, 2002. Dr. Lemke, a of the magnetization susceptibility long-time friend to CMRR, was at a reversal point. recognized for his “contributions to magnetic recording for computer data storage.”

CMRR REPORT7 WINTER•2002 CMRR WELCOMES NEW VISITORS

Yun Jun Tang from the Institute of Physics, Chinese Academy of Sciences will spend the next year with Prof. Ami Berkowitz’s research group. Dr. Tang received his Ph.D. from the Chinese Academy of Sciences in 1995/6, and was an Alexander von Humboldt fellow from 1997 - 1999 at Kaiserslautern University. He was a visiting scholar for one year, in 2000, at the Hong Kong University of Science and Technology before returning to the Chinese Academy of Sciences where he is an Associate Professor. Dr. Tang will be working on two programs in the Berkowitz laboratory. He will be carrying out the magnetic and structural characterization of the ferromagnetic- shape—memory-alloy (FSMA) particles that CMRR will be producing by spark erosion under the MURI-FSMA program. He will also be involved with the preparation and processing of these materials. The second program is an investigation of finite size effects in antiferromagnetic and ferromagnetic nanoparticles, which will be carried out in collaboration with Prof. Frances Hellman of the Physics Department. Dr. Tang will be responsible for the heat capacity measurements that are the heart of the investigation. The nanoparticles will be supplied from IBM Watson Research Center, Yorktown Heights, and Dr.Tang will be characterizing the magnetic and structural properties of these materials, as well as preparing suitable samples for the heat capac- ity measurements. Dr. Tang’s wife, Jing Xu, and son Owen will join him later this year.

Andreas Hanke and Matthias Panzer joined Prof. Frank Talke’s group in September. Both Andreas and Matthias received their pre-Diploma certifi- cate from the University of Rostock in 1999 and are enrolled as graduate students in mechanical engineering under the advisement of Prof. Christoph Worernle in collaboration with Prof. Talke. Andreas’ focus is on structural mechanics and engineering dynamics, while Matthias is specializing in engineering dynamics and vehicle engineering.

Eirik Rosnes Visiting scholar, from the University of Bergen, Norway, will be working jointly with Profs. Paul Siegel and Jack Wolf and their research groups through February 2002. Eirik has both a bach- elor’s and master’s in Physics from the University of Bergen and is currently enrolled in the Ph.D. program in the Department of Infomatics, under the supervision of Øyvind Ytrehus. Some of you may remember that Øyvind was a visiting scholar at CMRR from July 1998 to June 30, 1999. Eirik’s research will focus on parallel concate- nated convolutional codes (“turbo codes”) and high rate component codes. Besides his research, Eirik says that he likes to both play and watch soccer.

CMRR REPORT8 WINTER•2002 Alexandre Graell i Amat will be working jointly with Profs. Paul Siegel and Jack Wolf and their research groups through August 2002. Alex holds double master’s degrees in , one from the Universitat Politecnica de Catalunya and the other from the Politecnico di Torino. Alex’s research activities at CMRR will be related to the field of iterative decoding for magnetic recording channels. Alex is a Ph.D. student of Prof. Sergio Benedetto at Politecnico de Torino, Italy. Alex is quick to point out that although he is studying in Italy, he is in fact Spanish.

Open House

he University of Ryan Taylor, a graduate student Wire Recorder Radio Phonograph California, San Diego in Prof. Frank Talke’s group, pre- Combination, a circa 1905 Edison celebrated its 40th sented a lecture entitled “Design of Cylinder Phonograph, and a unique anniversary at the home- a Hard Disk Drive.” The audience Decca Portable Phonograph were coming celebration and learned that magnetic recording is given, followed by comparisons of Topen house on October 20 from now the dominant information stor- modern technology including R- 9:00 a.m. to 4:00 p.m. CMRR age medium in the world. Members DAT, DCC and the ubiquitous joined the celebration and enter- of the audience were asked to audio CD. The Majestic was donat- tained and educated visitors with remember that less than 10 years ed to CMRR by Sam Arn, and was the following activities. ago, if you wanted to look some- restored to near mint condition by Fred Jeffers, the Magnetic thing up in an old newspaper, you Dave Buck and daughter Heidi. Magician, presented his Magnetics would go to the public library and There were many other activities Magic Show in the CMRR auditori- look it up on microfiche film. In including: samples of “Scientific Ice um. Fred demonstrated real mag- 1956, IBM introduced the world’s Cream” (made by freezing the netic physics, doing seemingly first magnetic hard disk for data ingredients with liquid nitrogen) impossible things. Many of the storage. IBM’s first hard disk stored were served to visitors in the effects demonstrated are at first about 2,000 bits of data per square CMRR lobby, tours of the CMRR completely inexplicable, even to inch and had a purchase price of labs were provided, lots of posters experts. No slight of hand is about $10,000 per megabyte. By of CMRR’s on-going research activ- involved. Each trick is based on real 2001, storage capacity is nearing ities were on display, and an infor- physics, and many serve to intro- 100 billion bits of data per square mation table was set up on Library duce and “bring to life” important inch and the cost of storing a Walk to acquaint visitors with concepts in magnetics. The goal of megabyte has dropped to around CMRR’s activities and to point his talk is not only to entertain and one hundredth of one cent. Visitors guests in the right direction to join mystify, but also to teach. Unlike all were asked to have a look at some in the fun at CMRR. other magicians, this “Magnetic of the science used to advance this The open house was well attend- Magician” explained how each trick technology to its present state. ed campus-wide, and CMRR had works and outlined the basic CMRR Director Paul Siegel and approximately 100 visitors through- physics involved. Fred’s show is Prof. Jack Wolf gave demonstrations out the day, suitable for any audience, from the of recording equipment from past grade school level through the most ages. Demonstrations of a 1947 expert. Majestic Radio & Television Corp.

CMRR REPORT9 WINTER•2002 continued from page 4 detection or perhaps advanced high the storage hierarchy, bridging the One example is self-assembly of mag- sensitivity giant-magneto-resistance space between the fast but expensive netic nano-particles with length sensors.” DRAM and the massive and low- scales in the 3-10 nm range. Writing Several groups have proposed cost magnetic storage. The media and reading such small magnetic MEMS based probe storage devices could be non-magnetic, e.g. ferro- islands, however, will be extremely involving an x-y media stage and an electric or polymer materials. “At challenging. array of individually actuated probe Seagate Research, we are interested We need a lot of innovation. tips. Such devices could provide in all these areas,” added Weller. Eventually, some kind of probing faster access to data than convention- “Our philosophy is to place ‘some’ technology will be required: it could al rotating magnetic storage, yet pro- effort inside our own research labo- be based on spintronics effects, tun- vide large capacities at relatively low ratory, where researchers work on neling mechanisms, magnetic force cost. They would be intermediate in various approaches, materials and phenomena and stay in touch with what’s happening on the outside. The Wolf Group purchases a The further out we go, the more we have to rely on partners and acade- State-of-the-Art Guzik Tester mia.” Also making presentations at the The Guzik tester consists of two parts: the spinstand (S-1701B) and the September 21 session were two associated read/write electronics (RWA-2585S and ANA-985B). The spin- industry researchers who focused on stand is capable of micropositioning repeatability of greater than 0.3 MRAM development: NVE’s Jim microinches and selectable embedded servo for high TPI track following. Daughton, on “MRAM: Status and The electronics package includes a proprietary Guzik PRML channel for Opportunities”; and Motorola’s Brad data rates up to 1 Gb/sec as well as an adaptor board for accepting third- Engel, on “Prototype 256-kbit party commercial R/W channels. The system also includes a built-in digital MRAM.” The workshop concluded oscilloscope and spectrum analyzer. September 22 with presentations by Standard parametric magnetic recording tests are integrated into the Stuart Parkin, an IBM Fellow at the software in addition to advanced bit error rate tests through either the IBM Almaden Research Center, on Guzik or commercial R/W channel. Initially, the spinstand will be used for “Magnetic Tunneling Junctions and studying the noise characteristics of advanced heads and media as well as Magnetic Memory”; Joachim Ahner, measuring bit-error-rate versus signal-to-noise-ratio for various read chan- Seagate Technology, on “Novel nel designs with perpendicular magnetic recording systems. The heads and Imaging Technology and Potential media used with the spinstand are donated by CMRR sponsor companies. Multiple Probe Storage Devices”; and Ramamoorthy Ramesh, University of Maryland, on “Ferro- electric Memories: Science, Technology and Beyond.” After the workshop, organizer Frances Hellman was upbeat about the prospects of increased future cooperation between UC faculty, students and industrial partners. “We will now work on a research agenda to meet some of the chal- lenges outlined by industry researchers,” said Hellman. For more detailed information on the Workshop, including copies of the speakers’ presentations, please visit the Cal-(IT)2 website at: http://www.calit2.net/events/2001/nv m/9-21-nonvolatile.html JASON GOLDBERG, MEMBER OF THE WOLF GROUP

CMRR REPORT10 WINTER•2002 Obituary for Peter A. McEwen

eter A. McEwen, a University in 1990. leading expert in cod- Dr. McEwen joined ing and signal pro- Quantum P cessing for data stor- Corporation when it age, died Nov. 25, 2001 in acquired Digital’s Sunnyvale, California. He was 37. storage business in Dr. McEwen was manager of 1994. He returned to the Advanced Recording Channel school again at Architecture Group at Maxtor CMRR, receiving his Corporation in Milpitas, Ph.D. degree in California. His group was respon- 1999. He joined sible for evaluating coding, detec- Maxtor Corporation tion, error correction, and servo in 2001 as it merged algorithms and selecting the ones with Quantum’s disk to be used in read channels for drive business. next-generation products. He had Dr. McEwen’s the valuable capacity to integrate favorite topic was all the various coding for the magnetic recording details into an elegant read chan- channel. This was the subject for his A congenial colleague, Dr. nel architecture whose perform- dissertation, and he continued to McEwen collaborated with ance was excellent but whose show great interest in this area dozens of coauthors on research implementation was simple. throughout his career. His thesis papers, meticulously ensuring Dr. McEwen 's interests were included work on construction of they all received proper credit for diverse. He contributed to trellis codes and forbidden list codes. their contributions. He was research in the areas of detection Later he complemented these con- always willing to spend time theory, modulation codes, syn- structions with higher rate codes explaining codes and channel chronization, equalization, mitiga- such as event error control codes. architecture to coworkers. tion of thermal asperities, and in Dr. McEwen’s success was matching read channel architec- achieved despite being afflicted tures to specific head and media with scleroderma, a rare, chronic technologies. Dr. McEwen was autoimmune (arthritis) disease educated in electrical engineer- which impaired his ability to do ing, receiving his S.B. degree everyday activities such as writ- from the Massachusetts Institute ing and typing. of Technology in 1986. He went to work for Digital Equipment — Thomas Howell Corporation in 1986, returning to — Pablo Ziperovich school and receiving an M.S. — Ara Patapoutian degree from Carnegie Mellon

CMRR REPORT 11 WINTER•02 Current CMRR Sponsors Center for Magnetic Recording Research Reeporppoorrtt

Director: Agere Systems Paul H. Siegel Department of Defense/NSA Newsletter Editor: Hitachi, Ltd. (Central Research Laboratory) Cheryl Hacker Hitachi-Maxell, Ltd. Contributors: IBM Corporation Neal Bertram Jason Goldberg Imation Corporation Cheryl Hacker Infineon Technologies Gordon Hughes Doug Ramsey Maxtor Corporation Paul H. Siegel Quantum Corporation Jack Kiel Wolf Read-Rite Corporation Design & Layout Seagate Technology Sirious Design ST Microelectronics, srl Western Digital Corporation

University of California, San Diego First Class Mail Center for Magnetic Recording Research, 0401 U.S. Postage 9500 Gilman Drive PAID La Jolla, CA 92093-0401 La Jolla, CA http://www.ucsd.edu/cmrr Permit No. 128